|
Download README.md from Berkeley-ICON-Lab/SAMBAR-checkpoints: direct link, hf CLI and curl.
- Browser
- Download file 2.45 kB
-
https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/resolve/main/README.md
- Command line
-
hf download hf://Berkeley-ICON-Lab/SAMBAR-checkpoints/README.md
-
curl -L -H "Authorization: Bearer $HF_TOKEN" -o README.md https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/resolve/main/README.md
2.45 kB
| tags: | |
| - robotics | |
| - libero | |
| - jax | |
| - orbax | |
| # SAMBAR π0.5 LIBERO checkpoints | |
| Pretraining and finetuning checkpoints for [SAMBAR](https://github.com/labicon/SAMBAR). | |
| ## Pretraining checkpoints | |
| | Pretraining curriculum | Checkpoint directory | | |
| | --- | --- | | |
| | Narrow, 32 tasks | [libero-32/15000](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/libero-32/15000) | | |
| | Broad, 117 tasks (27 + 90) | [libero-117/9999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/libero-117/9999) | | |
| Each numbered directory contains Orbax parameters under `params/` and normalization statistics under `assets/`. The surrounding run folder includes the original hyperparameters and any logs provided with that run. | |
| To download the 32-task checkpoint after obtaining model access, run: | |
| ```bash | |
| hf download Berkeley-ICON-Lab/SAMBAR-checkpoints --include 'libero-32/15000/*' --local-dir ./checkpoints | |
| ``` | |
| For the 117-task checkpoint, use `--include 'libero-117/9999/*'`. Use the numbered directory as the pretrained checkpoint path. For sequential SAMBAR training, compute the initial LIBERO importance from the matching pretraining demonstrations using the [importance export instructions](https://github.com/labicon/SAMBAR/blob/main/docs/experiments.md#pretraining-and-initial-importance). | |
| ## Finetuning checkpoints | |
| These runs start from the 32-task LIBERO checkpoint. The sequential checkpoints are the final stage after tasks 3 → 8 → 11. Each independent run finetunes on one held-out task. The links point to the saved Orbax step directories. | |
| | Experiment | SAMBAR | SFT | | |
| | --- | --- | --- | | |
| | Sequential, tasks 3 → 8 → 11 | [9999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/sambar_sequential_finetuning/9999) | [9999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/sft_sequential_finetuning/9999) | | |
| | Independent, task 3 (pot-on-stove) | [19999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/narrow/task3_pot_on_stove/sambar/19999) | [9999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/narrow/task3_pot_on_stove/sft/9999) | | |
| | Independent, task 8 (bowl-in-drawer) | [19999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/narrow/task8_bowl_in_drawer/sambar/19999) | [9999](https://huggingface.co/Berkeley-ICON-Lab/SAMBAR-checkpoints/tree/main/narrow/task8_bowl_in_drawer/sft/9999) | | |